The influence of personalization on education and enjoyment in a museum

(1)

MASTER THESIS

THE INFLUENCE OF

PERSONALIZATION ON

EDUCATION AND

ENJOYMENT IN A

MUSEUM

L.M. Perloy

MSC HUMAN MEDIA INTERACTION

EXAMINATION COMMITTEE

Dr. E.M.A.G. van Dijk (University of Twente, Enschede, the Netherlands) F. van der Sluis MSc. (University of Twente, Enschede, the Netherlands) Prof. Dr. Ir. A. Nijholt (University of Twente, Enschede, the Netherlands) Drs. H.M.F. Kockelkorn (Museon, The Hague, the Netherlands)

DOCUMENT NUMBER

(2)

(3)

Abstract

Maintaining the museum integrity as a conserving and educational institution on one hand, and making the museum more competitive and popular on the other hand, is one of the main challenges for museums. This research tried to find out if a newly designed collaborative quest was suitable to be both educative and enjoyable. Special attention was focused on the influence of personal interest on education and enjoyment.

To find an answer, an experiment was conducted in Museon in which groups of four participants, with at least two children, played the quest. The quest started and ended at a multi-touch table, which was used because it was very suitable for collaboration. Between the parts at the table participants had to answer questions, which were provided at terminal computers in the main exhibition of Museon.

(4)

(5)

Preface

One of the things I find interesting about Human Media Interaction is using new technology to improve an existing situation. For this research I had the opportunity to work with one of the newest multi-touch tables on the market. I was involved in the development of a new interactive quest through Museon that made use of the table and the existing technological infrastructure. The quest was not only developed to find an answer to the research question in this thesis, but also to be available to all visitors of Museon.

I would like to thank the staff of Museon for all their support and especially my supervisor at Museon, Hub Kockelkorn, for his guidance and help in the development of the collaborative quest. Special thanks to Renske Koning and her family, who offered me a place to stay during the experiments.

Els Koster, Stefan Oldewarris and Alje van den Bosch were very supportive during the completion of this thesis. The first because of all her help regarding SPSS, the last two for their proof reading. Of course I would like to thank my supervisors at the University of Twente and especially Frans van der Sluis and Betsy van Dijk for their guidance and great support.

Finally I would like to thank my friends for the great time in Enschede and my parents and family for their support throughout my entire study.

Bert Perloy

(6)

(7)

Chapter 1

Introduction

“Museum. A museum is a non-profit, permanent institution in the service of society and its development, open to the public, which acquires, conserves, researches, communicates and exhibits the tangible and intangible heritage of humanity and its environment for the purposes of education, study and enjoyment.”

The quote above is from the International Council of Museums [1] and contains the definition of a museum. A museum has three goals: education, study and enjoyment.

Museums have to compete more and more with other educational; focusing on both education and entertainment, and cultural centers and activities [12]. One of the things museums have done to deal with this is making use of (recent) technology. They hope to make the museum visit more complete and more entertaining, and to increase the visitor numbers [23][28]. Making use of personal technology can personalize a museum visit, make it more interactive and give visitors immersive experiences. The accent of museums shifts more towards enjoyment, making the traditional museum more of a hybrid place, where education and recreation are both important [34][56].

1.1

Museon and PuppyIR

This research has been done for Museon, a science museum in the Hague, the Netherlands. Museon is an educational museum for a broad audience and the main themes are the human and his role in society, nature, culture, science and technology. Museon has 160.000 visitors each year, the majority of these visitors are families. These families are mostly composed of one or two adults and one or more children. The age of the visitors varies from four years old till above 80 years old.

Museon has one permanent exhibition, called Your World, My World. Visitors can discover for example how life came to be on Earth or how energy is stored. But also the relationship between Barbie and religion can be found [38]. In the exhibition space many terminals, about 120, are available. A terminal is a computer with a touch screen and a barcode reader. Visitors can scan their entrance ticket, on which a special code is printed, to get access to personal content. Appendix C shows a map of the museum.

Museon is a museum with the goals education, study and enjoyment, according to the given definition. It also faces the challenge of finding the right balance between education and enjoyment. The available terminals are currently only used for information provision. Together with a new multi-touch table, Museon wanted to use their technology with a surplus value for the visitors. Their wishes were that the focus was on collaboration and that the story behind the items in the collection of the museum should be told to the visitor. This research deals with this problem, with a focus on education and enjoyment.

1.1.1

Multi-touch table

(10)

doing, something that improves the communication. Because multiple concurrent input is detected, it is more likely that all users contribute to the interaction [48].

In earlier research, single- and multi-input devices have been compared. Stewart et al. [57] investi-gated cooperation behind a single screen with one or two input devices. A usability test with Kidpad, a multi-user drawing application, was conducted. Groups consisting of two children, aged 8-12 years, had to draw a picture with one shared mouse or two individual mice. Children found two input devices easier and more fun to work with. Fun was more important than efficiency of task completion. Children indicated that they preferred to work with multi-input devices because there was no turn-taking and they could work in parallel. This indicated that working together is easier when there are more input devices, like on a multi-touch table.

In Stewart et al. [58] the focus was again on the influence of multi-input devices on cooperation. This was tested with a system, KidPix. Two children had to draw a story using one mouse or two mice. The single-input variant caused more frustration and a lack of attention, and the quality of communication was less collaborative. In the multi-input variant the kids had more fun and collaboration was supported. These findings can likely be mapped to multi-touch tables.

Harris et al. [13] tested the difference in a design task on a table with single-input and a table with multi-input. The focus was on collaboration. Subjects, aged 7-10 years, had to design a seating plan for the classroom on a tabletop. Children in the first condition had to work on a single-touch table and children in the second condition on a multi-touch table. The condition did not affect the frequency or equity of interactions, but did have influence on the nature of children’s discussions. In the single-touch condition the children talked more about turn-taking. Also more frustration was observed. The multi-touch condition contained more task-related talk. The attention was more on the task itself, when using the multi-touch condition. Because of this, the multi-touch condition seemed to strengthen the relevant collaboration.

In the discussed papers multi-input devices were found to be good for collaboration, enjoyment and task-efficiency. This supports the initial assumption of Museon and makes the table suitable for the research.

1.1.2

The puppyIR project

The research in this thesis is part of an international research project, puppyIR [42]. When searching for information on the internet, most will use a search engine, like Google or Bing. But these search engines are developed for adults, and not always suitable for children [21]. PuppyIR is filling this gap, by, as can be found on their website, “constructing an Open Source Framework that will provide the infrastructure to develop ’child-focused’ and ’child-friendly’ components to be deployed within child information services” [41].

One aspect of the project is interaction. How can suitable information be presented to children, so that they can understand and work with it? Is a personal computer sufficient for children, or is there other technology that can be used? These are some questions that are still open and puppyIR is trying to find answers to them.

In this research project an implicit form of searching and a quite new technology, the multi-touch table, are used to make a contribution to the project.

1.2

Museum-going experience

As can be extracted from the definition of museum, education, study and enjoyment are important. Kotler and Kotler [24] observed that museum managers have to find a balance between those three. On one hand they want to make the museum more competitive and popular, on the other hand they want to maintain the museum’s integrity as a conserving and educational institution.

(11)

are outside the scope of the research in this thesis and are therefore not treated.

• Duration. The casual visitor spends around an hour in the museum [24, 51]. This average

spending time differs for each museum, but it gives an indication. The average time spent in Museon is higher, about two hours.

• More than spectators. Visitors want to be more than spectators. Museum visits can be made

memorable and unique by creating environments in which visitors can immerse themselves [24][36]. • Part of group/ collaboration. Museum visitors are mostly part of a group, like family or

friends. The behavior of this group is dependent on the interests and attitudes of the whole group [24]. Visitors find social and recreational experiences as important or even more important than intellectual and educational experiences [24].

• Personal relevance. Visitors have their own agendas, and visit a museum the way they want.

They prefer museum activities with which they can connect in an easy manner, activities that are relevant to themselves [24]. In Chapter 2 is explained how education and enjoyment are linked to personal interest and why it is important to make a museum visit personally relevant.

• Guiding. It can be hard to find one’s way in a museum for first-time visitors. To solve this problem,

large museums offer maps of galleries and exhibits [24]. Another problem is museum fatigue [7]. This is the phenomenon that visitors pay less interest in exhibits as the visit progresses.

All these concepts are to a greater or lesser extent important when improving the museum-going expe-rience. Personal relevance, guidance, immersion and collaboration are things a visitor finds important when visiting a museum. Study, enjoyment and education are the goals of the museum. Picture 1.1 gives a schematic overview of the relation between the goals of visitors and the goals of the museum. Improving the first four concepts improves the museum-going experience and helps a museum to reach their goals.

1.3

Collaborative quest

The main problem of this research project is that it is hard for a museum to find the right balance between education and enjoyment. Both education and enjoyment are influenced by personal interest, which is covered in more detail in Chapter 2. Taking education, enjoyment and personal interest into account, a collaborative quest was designed. This is an interactive guide through the permanent exhibition My World, Your World.

The quest consists of three parts. Visitors start with a group of three/four at a multi-touch table. They personalize the quest by making a selection from the exhibits of My World, Your World, by picking related images. This helps visitors to get an overview of the collection. With the selected exhibits as input, personalized routes through the exhibition are generated, giving visitors some guidance. This part will be referenced to as the begin part.

The second part, the middle part, is in the exhibition My World, Your World. A visitor scans his personal entrance ticket at one of the terminals. In response a question about one of the earlier selected exhibits is shown. The visitor can find the answer at that exhibit. Eventually all group members have answered their questions and return to the multi-touch table. In this last part, the end part, they can play a final game, that tests the knowledge gained by the group.

The focus of the research is on the design of the quest in general and the accompanying applications running on the multi-touch table. For the general design collaboration, education, enjoyment and per-sonal relevance are important. The target audience are visitors of Museon, who can play the quest in groups of three or four people.

1.4

Motivation and research question

(12)

Figure 1.1: Goals of visitor and museum linked to museum-going experience.

• Visitors can feel lost in a museum, the collaborative quest should give the visitors guidance, by providing them a route through the museum based on interest.

• Museum fatigue causes visitors to move faster through exhibits when they are longer in the museum. The quest helps them to divide their attention over the whole exhibition.

• Museum visitors are mostly part of a group. The collaborative quest should be suitable for groups, with a focus on collaboration.

• The use of the technological infrastructure, especially the terminals, of Museon is currently not optimal.

• The museum should be competitive with other cultural centers and activities, but education is still an important goal. Both education and enjoyment are linked to personal preferences. The collaborative quest should use the personal preferences (personalization) to create an educational and enjoyable activity, that is personally relevant.

Taking these motives into account, the research question is formulated:

“Can the educational and fun experience of a museum visit be enhanced by personaliza-tion?”

Requirements are the use of a multi-touch table, terminals and an accent on collaboration. The research question is very broad, so the following sub questions are used.

(13)

Besides these questions, three other questions are used to check the requirements. 3. Do people collaborate while they do the quest through the museum?

4. Are a multi-touch table and terminal computers suitable to be used in a quest through the museum? 5. Is the quest through the museum suitable to give guidance?

From this research question and sub questions four constructs are extracted: enjoyment, education, collaboration and personal interest. In Chapter 2 the theory behind these constructs is given.

1.5

Outline of research

(14)

(15)

Chapter 2

Theoretical framework

In this chapter theoretical background information about the four main constructs education, enjoyment, collaboration and personal interest is given. For each construct theoretical information is given and linked to the collaborative quest. The chapter concludes with hypotheses, which are drawn from the collected information.

2.1

Education

Education is one of the main goals of Museon. Originally museum education was limited to ’taught’ lessons for school children and tour groups. Those are still important, but museums can do more for education. Exhibitions, displays, events and workshops can be used to educate visitors [17]. The educative material that is available in Museon is the collection in the exhibit My World, Your World, so in this case the exhibition is used for education.

This section looks at age-related limitations, the three major learning theories and the three domains of educational activities. The information obtained from the theories is used to design the educational part of the quest.

2.1.1

Age-related limitations

The target audience of the quest are families with children. Jean Piaget, the Swiss psychologist, believed that certain ways of thinking were more difficult for children than for adults. According to Piaget, people pass through four different stages, the four stages of cognitive development. The first stage is the sensorimotor stage (0-2 years), the second is the preoperational stage (2-7 years), the third the concrete operational stage (7-11 years) and the last one the formal operational stage (11-adult). In the first two stages, children are able to think operations through logically in one direction and have difficulties seeing another person’s point of view.

In the concrete operational stage, children are able to solve concrete problems in a logical fashion, understand laws of conversation, are able to classify and seriate, and understand reversibility. In the formal operational stage children become more scientific in thinking, are able to solve abstract problems in logical fashion and develop concerns about social issues.

One limitation of Piaget’s theory is the strict divisions between the stages. The thinking process is continuous and not the same for everyone. It also appears that Piaget underestimated the cognitive abilities of children and overlooked cultural factors. Psychologists agree that children go through the changes described in the stages, but it is not a perfect guide [62].

2.1.2

Learning theories

(16)

Behavioural learning Behavioural learning is a learning type where external events are seen as the

cause of changes in observable behaviours [62]. Several non-exclusive views on behavioural learning exist. One of these views is operant conditioning, developed by Skinner. Operant conditioning is about reinforcements and voluntary behaviour. Antecedents are events that precede an action and consequences are events that follow an action. From the relation antecedent - behaviour - consequence people can learn. If you are for example tired (antecedent), you can choose to sleep (behaviour). This might result in feeling awake afterwards (consequence). From this consequence you have learned that it helps to sleep when you are tired. Consequences can be used to strengthen or weaken behaviour. This is called reinforcement. In positive and negative reinforcement something positive is added or something negative is removed. Punishment is the opposite. Something negative is added or something positive is removed. The first form, adding something negative, is very effective, but there are doubts if it is a good thing to do [11]. One of the problems is generalization. Children can associate other stimuli with the punishment stimulus and make the same response to these associated stimuli.

Another view is observational learning [62]. Albert Bandura thought that earlier learning views were accurate, but gave only a partial explanation of learning. Especially social influences lacked in the traditional views. He developed a new theory, the social learning theory. In this theory people can learn by observing other people. An extension of the social learning theory is the social cognitive theory. It adds cognitive factors as beliefs, expectations and self-perception. There are two variants: enactive and vicarious learning. The first one is like operant conditioning. Where in the operant conditioning theory consequences reinforce behaviour, in enactive learning, consequences provide useful information that can be used for future decision making. Vicarious learning is learning by observing other persons being reinforced for behaviour.

Cognitive view of learning The second theory is a cognitive view of learning. In contrast with

behavioral learning, where people learn new behaviours from responding, the cognitive view is about the people themselves, who are constructing their own learning.

Learning is seen as an active mental process where knowledge is acquired, remembered and finally used. With new knowledge it is possible to change behaviour. The difference can also be seen in the reinforcers. In behaviourism they strengthen or weaken the responses, in cognitivism they serve as feedback about what is likely to happen when something will be done again or avoided. So a reinforcer provides new information that can be used for future decisions.

Cognitivism makes use of schemas. Knowledge is stored in schemas, internal knowledge structures. When someone receives new information, the information will be compared to existing schemas. The schema tells the person what he can expect from a situation or object. An existing schema can be combined, altered or extended to accommodate new information [35]. Coherent to this is the concept of meaningful effects. People remember and learn information that is meaningful to them easier.

Constructivist view of learning The last theory is the constructivist view of learning [62]. It has

its roots in the cognitive view of learning, but has moved beyond. Constructivism is about the personal and active construction of knowledge. A person has his or her own perceptions and experiences. These are interpreted and with the interpretation an own reality is constructed. So prior experiences, beliefs and mental structures that are used to interpret objects and events determine ones knowledge. This is done actively [35]. The process of social interactions shares the realities of different persons with each other. This emphasizes the importance of collaboration.

For the construction of knowledge some things are important. Marlowe and Canestrari [31] pointed out the difference between shallow and deep knowledge. With shallow knowledge only conclusions or facts are remembered, not the connection between those two.

2.1.3

Domains of educational activities

There are three domains of educational activities [6, 32]. The domains are cognitive, affective and psychomotor. The cognitive domain is about mental skills and often referred as the Knowledge domain. Affective is about growth in emotional areas and feelings (Attitude domain) and psychomotor about physical or manual skills (Skills domain). Attitude and Skills are outside the scope of this research.

(17)

Figure 2.1: Bloom’s Taxonomy of the Cognitive Domain. The first two levels were used in this research. Lorin Anderson [10, 25]. The new version, which is showed in Figure 2.1, is used in this research. The important categories for this research are remembering andunderstanding. Remembering is about

retrieving, recognizing, and recalling relevant knowledge from long-term memory. Understanding is

about constructing meaning from oral, written, and graphic messages through interpreting, exemplifying, classifying, summarizing, inferring, comparing, and explaining.

2.1.4

Education in the quest

Based on the four stages of cognitive development, the quest is designed to be suitable for children older than seven years of age, but this is not a strict limit. Questions in the quest are not suitable for children younger than seven years, because concrete operational stage skills like classification and seriation are required. In the Netherlands children learn to read and understand sentences from the age of seven [22]. That is also important because participants should be able to read their questions without help.

The quest uses the principle of reinforcement. At the start of the game participants are told to have fun and to do their best so they might get a high score. This is mainly a performance goal, which is not the best goal for education. Mastery goals are better, but harder to use in this setting [62]. During the quest each participant has an individual score and there is a common group score. An incorrect answer is punished with a lower individual and group score, a correct answer rewarded with a higher individual and group score. The group score is used to stimulate cooperative learning. This score reminds the group members that they are linked with each other. If they coordinate their efforts, the likelihood to accomplish the task (getting a high score) increases. People can learn by watching others, for interaction and content purposes. The multi-touch table helps with this, because one of the strong points of the table is simultaneous interaction.

After each answer feedback is given. The participant is told why the answer was correct or incorrect. Feedback on incorrect answers helps visitors to remember correct information, because otherwise incorrect information can be remembered as correct [31]. With this extra background information the quest provides information that makes it possible to obtain deep knowledge and not only shallow knowledge. This helps the users to understand the information (understanding) and not only remember the factual

information (remembering). The stimulation of finding the answer in the exhibition also helps. So

(18)

From cognitivism it is expected that a personalized quest has a positive effect on learning. This is also supported in Renninger et al. [46]. Children and adults with an interest in an activity or a topic persist for longer periods of time, pay more attention, and acquire more knowledge than children or adults without that interest.

2.2

Enjoyment

One of the goals of a museum is enjoyment. There is no single definition of enjoyment and it is often interchangeably used with terms like pleasure and fun. In Blythe and Hassenzahl [3] the differences between enjoyment, fun and pleasure are described. Concluded is that none of the terms are reducible to single definitions and that fun and pleasure are two distinct types of enjoyment. Fun is about distraction from one selves, from ones concerns and problems. Pleasure is deeper, it focuses on an activity and gives a feeling of absorption. So fun can be used to make repetitive and routine work more enjoyable. Pleasure, with its focus on absorption, can make non-routine and creative work enjoyable.

Lin et al. [29] gave an overview of several definitions of enjoyment. In psychology for example Davis expressed the view that “A is enjoying E, if E is causing A to have a number of occurrent beliefs concerning E, which collectively add significantly to the pleasure (happiness) A is experiencing”. They conclude that all the definitions provide a basis for the following definition of enjoyment: “The meeting and fulfillment of a person’s needs.” This is the definition that is used for this research.

Vlachopoulos and Karageorghis [61] mentioned that feelings of enjoyment are a consequence of in-trinsically motivated behaviour. The connection between enjoyment and both intrinsic and extrinsic motivation is also made in Reeve [45]. It focuses mainly on the difference between enjoyment and in-terest in intrinsic motivation. The conclusion is that enjoyment contributes to intrinsic motivation by sustaining the willingness to continue and persist in an activity. This is the exact opposite of what was said in Vlachopoulos and Karageorghis [61], where enjoyment was a consequence of intrinsically moti-vated behaviour. Notwithstanding that, there seems to be a strong connection between motivation and enjoyment, so more information on motivation is given below.

2.2.1

Motivation

The definition of motivation given in Woolfolk et al. [62] is “An internal state that arouses, directs and maintains behaviour.” An important aspect of motivation is the positive relation with time spent in activities. Motivation can be intrinsic and extrinsic.

Intrinsic motivation is associated with activities that are their own reward. The activity itself is

in-teresting or enjoyable, so people are motivated to engage in the activity for its own sake [20]. Several approaches to intrinsic motivation exist [62]. The humanistic approach to motivation was one of the first theories that explained why people act as they do. The approach is about encouraging people’s inner resources. A person aims at successful personal development and needs self-determination. The need for competence, control and relatedness are basic in people. This can be used in teaching. Expectations should be clear and achievable, and choices help increasing the perception of control and autonomy. Per-sonalized examples are good for relatedness. The sociocultural view of motivation focuses on the concept of identity. People find their identities and interpersonal relations within a community important. They participate in activities to maintain those identities and relations. For example, people are motivated to learn if they are at a school that values learning [62].

Extrinsic motivation is motivation created by external factors like rewards and punishments. Someone

(19)

tell that positive feedback concerning success actually increases intrinsic motivation and self-reported enjoyment. Saying things like “good job” can be dangerous [31]. It is better to say nothing, say what you saw or ask questions. When using praise it should be about the effort that children put into a task and not about their intelligence.

Several motivational techniques for classroom teaching are mentioned in Gage and Berliner [11]. It is important to give children a reason to be motivated. This can be done by telling what they are going to do and why it is interesting and enjoyable. It is also important to give them a clear goal. An example of where this went wrong was in an exhibition of the Berlin Museum of Natural History [18]. The museum accommodates the Tree of Life table. This is a 75 cm high table, with a 1.15 by 2.15 meter surface, on which an image is projected. Using capacitive sensor technology, the table is suitable for multi-touch. Four species related questions are displayed on the table. When touching them, the answer appears. The table can be seen as a sort of interactive information browser. One of the main critics on this exhibition was that the purpose was not clear. Without knowing the purpose, people are missing an incentive to persist in interaction [31]. Another thing that can be motivational is the use of familiar material for examples. Familiarity, associations and meaningfulness can improve retention and learning. A last point is to use what children have learned before.

2.2.2

Flow

Sweeter and Wyeth [59] described a research about enjoyment conducted by Csikszentmihalyi. He asked over a thousand respondents why experiences are enjoyable and found that optimal experience, called flow, is everywhere on the globe the same. The participants in his research described the different activities they found enjoying, in the same ways. One of his conclusions was that enjoyment (or more specific pleasure if the distinction from Blyth and Hassenzahl [3] is used) is the same regardless of age, gender or social class. He described flow as “an experience so gratifying that people are willing to do it for its own sake, with little concern for what they will get out of it, even when it is difficult or dangerous”. Flow experiences are composed of several elements, which when combined give a sense of deep en-joyment. The first element is that the task can be completed. The task should also have clear goals and should provide immediate feedback. People should be able to concentrate on the task and exercise a sense of control over actions. The last three elements are about immersion. In flow people experience a deep but effortless involvement that helps them forget the frustrations of everyday life. The concern for self disappears during the activity and the sense of duration of time is altered. When an activity fulfils these elements, people experience it as enjoyable.

2.2.3

Enjoyment in the quest

There does not exist a single manual that describes how an activity can be made enjoyable. People who are intrinsically motivated will engage in an activity. Giving users familiar examples, choices, and clear and achievable expectations helps people to become intrinsically motivated. The quest provides those by letting the user choose their own subjects and by giving a clear and reachable goal.

(20)

for answers in the exhibition. Finally they can show their team members what they have learned. By keeping users busy all the time and let them explore the museum, the quest should become immersive and together with the fulfillment of the other elements of flow, enjoyable.

2.3

Collaboration

The learning theories implied that working together has benefits for learning. In Woolfolk et al. [62] three variants of working together are mentioned; collaboration, cooperative learning and group work. And although collaboration, cooperation and group work are often used interchangeably, they are different concepts. Cooperation can be viewed as a way of learning by talking and working together to share thoughts. Collaboration is agreed behaviour contributing towards reaching a shared goal. When people work together, it is called group work. In group work people can collaborate or cooperate, but they don not have to. So for example, a group of five people is allocated a task. They do not work together and one person does all the work, but it is still group work. When they discuss the task together to solve the problem they cooperate. They can also divide the task in subtasks, work independent on them and assemble the outcome. This is collaboration.

Collaboration is the main concept for the collaborative quest. A team starts with the common goal of getting a good group score, but the individual group members have their own tasks. They can cooperate, but they do not have to. Collaboration is necessary to reach the common goal. For the parts at the table the group has to cooperate. The rest of the section treats cooperative learning, which shows the advantages of cooperation in the area of learning. After that several examples are given of how technology in museum helps stimulating collaboration.

2.3.1

Cooperative learning

Gage and Berliner [11] support the statement that working together has benefits for learning. They say that cooperative learning has a positive effect in learning. In cooperative learning small groups of students work together for group rewards. One of the advantages is that higher-ability children can help lower-ability children, helping each other with learning. The higher-ability children, however, should not be restricted by this. With one common goal, group members can help each other, but each individual should also be challenged on his own level. The positive effect of better involvement was demonstrated in Cappelletti et al. [4]. Children got the task to create a story together. Before the task, they had to tell an individual story, for comparison. The result was that the cooperation had a positive effect on involvement. However, the quality of work scored lower. This was probably because it is harder to create a story together than alone.

In Cappelletti et al. [4] several main tenets are given for cooperative learning. Positive interdepen-dence must be established among children. The result of this is that children feel linked with their group mates. They can only succeed if their group mates do, they have to coordinate their efforts with the efforts of their group mates. When done well, they can accomplish the task. Positive interdependence itself consists of several facets. With positive goal interdependence the group and individual group mem-bers have the same goal. In positive resource interdependence the focus is on resources. Only limited resources are available and have to be shared with the whole group. Individual performance can be pro-vided, by showing for example a score, in order to stimulate the individual accountability and personal responsibility.

2.3.2

Technology and museum exhibits

(21)

visible to others. Though it is possible to interact simultaneously, this was not done in this exhibit. People watched others interact with the painting, they are called public. After the current user moved on, someone from the group of public became the next user.

Peltonen et al. [40] showed that it is also possible that strangers interact simultaneously, but on their own private spaces. In Helsinki a vertical 2.5 meters wide multi-touch screen was installed in a shop window. The screen could be used from outside. On the screen photos could be viewed, rotated and resized. The screen was mostly used by more than one person, both collectively and parallel. With parallel use, users had their own private space and tried not to disturb other users. Collective use was using the whole screen with a group of people. Making the interaction visible to others is positive for learning. People can see that it is possible to interact and how things work, by watching others. Collaboration between people and their creativity led to unexpected behaviour. Instead of using the application as a photo viewer, people played the game Pong by throwing photos to each other.

Deus Oculi and Ghostship [15] are two exhibits that need collaboration to discover the full potential. Deus Oculi, exhibited at Chelsea Crafts Fait consisted of a main picture that displayed a Renaissance scene. On both sides are two fake “hand-mirrors” located. The Renaissance scene is made of two people, a man and a woman. On the places of their heads are holes, with monitors behind each of them. Cameras hidden in the hand-mirrors sent images to the monitors. If someone is looking at the mirror, his head is embedded in the painting in the middle. The exhibit provoked a lot of interaction and discussion between visitors. The same thing was done for Ghostship, but this time on a larger scale. People needed other visitors to find out all the options of the exhibit, with only two or three people it was hard to discover everything.

The setting and surroundings of an exhibit can help to stimulate collaboration. In Nottingham Castle, exhibits Storytent and Sandpit were designed to support collaboration [2]. Storytent was a place in the shape of an A-frame tent. Inside the tent, visitors had to place paper clues on a turntable. After placing the papers on the table, the display was triggered showing a 3D historical reconstruction. By rotating the turntable, a 3D panorama and related paintings could be viewed. Inside the tent multiple seats were placed to support interaction with more than one person. For Sandpit the same thing was done. Sandpit is a virtual sandpit, projected on the floor. By shining flashlights on it, users could dig for images. By placing paper clues in a nearby sandbox, the images in the sandpit changed. With multiple flashlights, the exhibits intention was also collaboration between visitors.

2.3.3

Collaboration in the quest

Collaboration is stimulated directly at the start of the quest. First of all, the quest is only suitable for three or four players, but that can make it just group work, without any collaborative or cooperative elements. Next the multi-touch tables physical shape requires the users to stand around the table. Not all the explanation texts on the screen are offered to all group members. In most cases there is only a single instance. The users can choose to read the text one by one, but the idea is that they help each other and one team member reads the text aloud. Around the table is enough space for public, who can join a group or just can see what the other group is doing.

The concept of the quest stimulates collaboration, because a common goal is given where the knowl-edge and skills of all users are necessary for the final result. Incorrect answers not only influence the individual score, but also the group score. To get a good team score, efforts have to be coordinated and collaboration can be helpful. In the final game the knowledge of all team members is needed to get the best result.

2.4

Personal interest

Personal interest is highly related to enjoyment and motivation. Reeve [45] described that interest and enjoyment both have differential contributions to (intrinsic) motivation. Interest helps people to start with an activity. It arouses the initiation and direction of attention and exploratory behaviour. After that enjoyment helps people to continue and persist in the activity.

(22)

environment generate situational interest. The consequence is an immediate affective reaction, that can last, but does not have to. Where personal interest is linked to enjoyment and liking, situational interest is not necessarily linked to positive feelings. For example someone can find a tarantula interesting, but does not like the spider. But when situational interest persists, also positive emotions can arise. According to Hidi and Harackiewicz [14] several investigations have shown some differences between children and adults who are interested in specific activities and children and adults who are not. The first group persists longer in the activity, pays closer attention, learns more and their enjoyment is greater.

In Hsi and Fait [19] an example of a museum is given, that made the visit more personal by making use of personal physical items. Each visitor got a personal RFID card, which they can use at several exhibits. At the exhibit “Heat Camera” visitors were able to see thermal images of themselves. With their RFID card, they could trigger a camera to take a picture of the scene. With their ID card number, they had access to these pictures, at the museum or at home. Concluding, identification technology was used to identify visitors and offer them personal content.

2.4.1

Personal interest in the quest

All the participants get a personal identification ticket at the start of the quest. The ticket links the table to the terminals in the exhibition. Each group member has to choose his/her own, personal items at the table. They can choose between items that represent parts of the exhibition. With the items a personal route is generated, that correlates to the items the visitor found most interesting. The personal ticket is used to get access to individual question at the terminals.

2.5

Hypotheses

In chapter 1 the main research question and related sub questions are defined. Based on the theory in this chapter, the following hypotheses are presented.

Sub question 1: Is there a noticeable influence of personal interest on the enjoyment of museum visitors?

Two variants for the quest are developed. One uses selected items, the other one uses the remaining items. According to the theory enjoyment is greater in activities people are interested in than in activities they are not. This leads to the following hypothesis:

H1: The reported enjoyment will be higher for the variant that takes personal interest into account than

for the one that does not.

Sub question 2: Is there a noticeable influence of personal interest on the education of museum visitors?

People remember and learn information that is meaningful to them easier and learn more from activities they are interested in. Expected is that visitors learn more from the variant that uses the personal interest than from the other variant. This leads to the following hypothesis:

H2: Visitors learn more from the variant that takes personal interest into account than from the one

that does not.

For the other three sub questions no hypotheses are drawn. Those questions will be answered using qualitative data.

(23)

Chapter 3

Development and implementation

To find out if the concept of the collaborative quest worked and to get answers on the research questions, an experiment was conducted. For this experiment a prototype of the quest was implemented. This chapter describes the final implementation of the prototype. Research to several aspects of multi-touch is conducted and the first part of this chapter gives an overview of this. The second part of the chapter describes the prototype, which took into account the findings of the multi-touch literature.

3.1

Multi-touch guidelines

Multi-touch tables are a relatively new technology. Unlike for example mobile websites, there are no clear design guides for multi-touch tables. But there is quite some literature written about experiments using multi-touch tables. This section summarizes the recommendations from that literature. The results are used as guideline for the quest. In Section 3.3 the guidelines are linked to the prototype.

Workspaces On multi-touch tables multiple users interact simultaneously. In Rick et al. [47] research

was done to find out how children work together at a multi-touch table. An experiment was conducted whereby children had to allocate where students in a class should sit. On each table there was space for three students and there were several restrictions. The children could drag the students to a table, allocate them to that position. One of the important findings was that children work mainly in the area in front of them. Children positioned at the left short side of the table did less, because they are often right-handed and get in the way of the children positioned right of them. In contrast to adults, children used the whole surface of the table, they touched all of the table. Also arms were freely moved across arms of other children, this in contrast with adults.

In Scott et al. [54] two observational studies have been conducted to gain more understanding of natural interaction practices involving tabletop interfaces. Traditional media like pen and paper were used in the studies, and with the results relevant design implications were drawn. The first design implication was the provision of visibility and transparency of action. This can be important if additional smaller screens for private use are available. The second implication was that every user should be provided with appropriate table space, which supports the findings in Rick et al. [47]. With individual tasks, every user should have enough space, but with a collaborative task, a small individual workspace and with that a larger collaborative space, can be better. A last useful implication was about functionality that should be provided in appropriate locality. A button that affects all, should for example be placed in the center. This way every user has the opportunity to participate in the decision of pushing the button.

Orientation One of the challenges in designing for multi-touch tables is orientation. Unlike a computer

(24)

solve this. Environment-based automatic orientation rotates items in the direction of the closest edge. The assumption is that the person who is closest to the item, should have the best view on it. In the paper these and other approaches of orientation are investigated. With the observational data, several implications for design are given. The first one is that users should always have the option to rotate items freely. When items are automatically rotated, the user should have the option to override this and place the item according to their wishes. The next one is that when a user has positioned an item, that orientation must be maintained. The last one mentions that rotation actions must have clear feedthrough; it must be clear to other users when a user is performing a rotation action.

Ryall et al. [50] supports that there is no single solution for orientation. When users are around the table, they have different views on the surface. For small chunks of text this was not a problem, for large amounts of text orientation was more important.

Multiple users With multiple users who can interact simultaneously, it can be helpful to use

coordi-nation schemes. Morris et al. [37] researched multi-user coordicoordi-nation at a tabletop. The main focus is on working with documents, but some conclusions are still interesting. Several global coordination policies were tested, from which voting was the most interesting one. If there is a global change on the table, a change that affects all the users, each group member can vote against or in favour of the change. Voting can be used to give all members an equal vote. There were also some policies for element coordination. Element coordination is about elements displayed on the table, which can be claimed by more than one person. In the public variant there are no restrictions, everyone can access an element. The stalemate variant makes an element temporarily inactive if someone attempts to take an element from someone else. The last variant is speed and force. The user who applies the most speed or pressure on an element, will get the element. One conclusion was that elements on a walk-up and use table in a museum, like the multi-touch table in Museon, should not have a policy. Controlling access is not desirable in this setting because none of the table elements belong to a specific user.

Ryall et al. [49] conducted an experiment to find out what the influence of table size and group size are. Participants were given a poem on paper. This poem had to be reconstructed on a multi-touch table, making use of single words. The results were that the size of the table had no significant effect on speed, but group size had a significant effect on speed. Larger groups were faster in their task than smaller groups. It was tested with groups of two, three and four people. Other interesting findings were that participants did not grab words that were near a partner. For large groups it can be interesting to extend the surface with additional vertical displays, where shared information can be displayed on.

Interaction Ryall et al. [50] have observed four interactive tables in non-controlled settings. The

table they used was the DiamondTouch. The first setting was a lobby table and several games were installed on the table. In the next setting a table was used to support bio-diversity researchers with a flora and fauna browser application. The third setting was a table on a three-day conference, NextFest 2004. Several games were installed at this table. The last setting was a two-day symposium, organized by the National GeoSpatial Agency. A map application was installed. All of the tables were observed and some usage patterns were noticed. Surprising was that these patterns occurred at most or at all of the four tables, despite the varied user populations and applications.

Subjects were hesitant to touch the table at the same time. This is more true for adults than children. When users interact simultaneously, they are concerned of touching each other. Another important finding is the ambiguous touch. The table registers every touch, even unwanted ones. Pointing and accidental touches, like leaning with wrist or elbows, are seen as input. An edge around the table can help with this. Users also prefer one-finger interaction. They rarely attempt to perform gestures on tabletops. Some basic finding is the finger resolution. Peoples fingers have different sizes. Because fingers cover elements when they touch it, for menus it is desirable to display labels slightly offset. The elements should also be large enough.

(25)

Epps et al. [9] researched user preferences for tabletop gesture interaction. On a large, rear-projected horizontal screen a static image was displayed. To find out which gestures users prefer, subjects were asked to perform a specific action on one or more of the objects displayed in the image. Requested actions were for example selecting, moving, opening, rotating, scrolling or zooming. Findings were that the index finger was by far the most used hand shape. A flat hand with fingers apart, a flat hand with fingers together, grabbing/releasing and a vertical hand shape were used commonly.

In Rick et al. [48] one very interesting observation was made. Children, even older ones, lift their fingers while they drag digital objects. The table can see this as a release action and drop the objects in the wrong place. Children can easily recover from these mistakes, but it can be frustrating.

3.1.1

Guidelines for the quest

Concluded from the literature discussed above, the following multi-touch table (MT) guidelines are used for the quest.

• MT1: Give users a private workspace in front of them. • MT2: Place global items in the center of the table.

• MT3: Do not require children to drag items over a long distance, without the option to let them lift their fingers.

• MT4: Make clear that the table can be used simultaneously.

• MT5: Use a physical border around the table, so that users can lean without touching. For tables without physical border, be careful with using the edges of the table.

• MT6: Take accidental touches into account. An accidental touch should not have a lasting effect. • MT7: Make items suitable for different sizes fingers.

• MT8: Support users in orientation, but give them the opportunity to override.

• MT9: Make everything suitable for interaction with a single finger, and for interaction with multiple fingers.

• MT10: For important group decisions, use a central voting mechanism.

These guidelines should help the user to interact in a natural way with the multi-touch table.

3.2

Prototype

With the multi-touch guideline and the theoretical framework the initial idea of the collaborative quest was fine-tuned and a first prototype was developed. The prototype was developed for use in the experi-ment.

3.2.1

Hardware

The multi-touch table that has been used is the MultiTouch Cell 46” from the company MultiTouch Ltd, located in Finland. The MultiTouch Cell has an 46 inch (117 cm) TFT Display, with a resolution of 1920 by 1080 pixels and an active area of 1017 mm by 573 mm. The table itself measures 1052 mm (width) by 605 mm (height) by 630 mm (depth). The table was placed on a wooden frame of circa 20 cm high, which positioned the screen circa 80 cm above the ground. The table can recognize markers with a minimum size of 90 mm by 90 mm.

The terminals in the museum are single-touch screens placed in wooden frames. Besides the touch input, users can scan their entrance tickets using the barcode readers. The screen size of the terminal screens is 15 inch (38 cm) with a resolution of 1024 by 768 pixels.

(26)

(a) Barcode for terminal.

(b) 2D barcode for table.

Figure 3.1: The markers.

3.2.2

Software

The software developed for the quest consists of two parts. The software for the multi-touch table and the software for the terminals. The two are connected via a central database that can be reached using web services.

3.2.2.1 Terminal

The terminal software is developed by Atos Origin. It is written in HTML, JavaScript and CSS. When users scan their entrance ticket and they have started the quest at the table, the first Collaborative Quest screen is displayed, as in Figure 3.2. The screen gives an explanation of the quest and some tips. The top bar is always visible. With the close button in the right upper corner the user can close the current screen. The terminal then returns to the standard welcome screen. If the user does not touch the screen for a while, the terminal automatically returns to the welcome screen.

Next is the question screen, shown in Figure 3.3. The prototype uses five question types. Example questions can be found in Appendix D.

1. Multiple choice. The user must pick one answer from multiple options.

2. Multiple response. The user must select one or more correct answers from multiple options.

3. Sort object order. The user must arrange the elements in the correct order.

4. Point & click. The user must touch the correct part of an image.

5. Match. The elements at the right side on the screen must be dragged to the corresponding labels

(27)

Figure 3.2: Terminal with welcome screen.

An optional small introduction is given, followed by the question. Below the question the answer options are displayed. At the right side of the question an image related to the question is shown. If the image does not make clear where to find the answer in the exhibition, an additional help button below the image is displayed. When the user touches the help button, the exhibit where the answer can be found is shown. The bottom of the screen contains several interface elements. First the individual score and the group score. Next a progress bar, which consists of a bar for each question. When a question has been answered, the bar becomes orange. A question cannot be answered immediately, so the user is encouraged to explore the museum to find the answer. After a rescan the question can be answered. When the answer is incorrect, the selected option becomes orange. If it is correct, it becomes green and some extra background information about the subject is shown. After three incorrect attempts, the correct answer is given.

Next is the item screen, displayed in Figure 3.4. After each question the user has to pick one item out of three. All items are related to the question and the exhibit. The items are used to connect the question with the final game. Below each item a description about the item can be found. The last terminal screen tells the user he or she finished the quest and should return to the table. If some of the other group members are not finished yet, a suggestion is also given that they could possibly be helped.

3.2.2.2 Multi-touch table

The software for the table is written in Java. It uses the open source multi-touch Java framework Multitouch for Java (MT4j) [27]. In Figure 3.5 the structure of the multi-touch table software is shown. The software consists of several scenes. Two screens are available without logging in. One is the start screen, where a group can be formed. In the other screen users can select a language. It is not possible to switch the language once a new game has been started.

Select language The language screen is one of the screens that is available to all users. The main

(28)

Figure 3.3: Terminal screen with a question.

table, direct feedback is given in the form of a painting. Each finger serves as a paint brush and all users can paint with as many fingers as they want. This way of interaction should be fun and should make the user aware of the fact that they can touch the table simultaneous. An example of the language screen can be seen in Figure 3.6.

Group scene The group screen is the start of the collaborative quest. In Figure 3.7 a screenshot of

the scene is shown. Users can form a group. A person has to put his marker on one of the four ticket places, each in the middle of a table edge. When a marker is placed, a slider appears. The knot of the slider can be moved to the circle in the center. The knot and part of the circle turn green, this way the user can see that he has joined the quest. By pressing the button in the middle of the center circle the group can confirm that it is complete. This is done to deal with accidental touches and to let every group member have a share in the decision process to start the quest.

After the users have logged in, four scenarios are possible. If none of the group members are in a quest, a new quest is started and a screen is shown with an explanation of the quest. If all group members are finished, the group is redirected to the final game. It is also possible that not all group members are finished with the middle part of the quest. A warning is shown that some group members are still busy and might need some help. The last scenario is that the group is incorrect. This can be the case when only three of the four existing group members log in. Or when users of different groups log in, trying to create a new group when the old groups not have finished their quest. A warning will be displayed.

Similar to the language scene, feedback is given on every touch at the table. When the table is touched, a particle effect with a glow is shown. Again to make it fun and to show that simultaneous interaction is possible.

Explanation scene When a new quest is started, the first screen the users see is a screen with a

(29)

Figure 3.4: Terminal item choice screen.

what they can expect from the quest. The explanation is displayed on a virtual page in the middle of the screen. This page can be resized by placing two fingers on it and move them towards or away from each other. The page can also be rotated. This can be done by placing two fingers on the page and make a rotating movement. A book icon is placed next to each marker. When this icon is touched, the page with the explanation is automatically rotated towards the corresponding marker. To stimulate the collaboration, only one page with an explanatory text is used. Users need each other because the text is not readable from all sides of the table. By pressing a green arrow, the group can proceed to the next screen, where they can write their names.

Account scene To make the quest more personal, users have to write their names on a Scrabble-like

screen. Figure 3.9 shows this screen. The names are used to link the marker to an account in the MuseOnline system, the system that is running on the terminals. Without a personal account it is not possible to get personal content at the terminals, like the quest. Stacks of letters are placed in the middle of the screen. The user can drag a letter to his personal rack. By repeating this, a name can be written. When the user has finished his or her name, he or she can confirm that by using the slider. When all users have confirmed that they are finished, the system proceeds to the preferences screen.

Preferences scene The preferences scene is an important screen. Users can select items that determine

their route through the museum. These items are displayed as circles with an image inside them, as can be seen in Figure 3.10. Each item is duplicated for the amount of users, so that each user can pick the same item. Six item slots are located around the marker. Users can drag the items in those slots to select them. When all six slots are filled, they can confirm that they are finished by dragging the slider. If all users have confirmed that they are finished, a screen with an explanation of the middle part of the quest is shown.

(30)

exhi-Figure 3.5: Structure of multi-touch software.

bition My World, Your World is necessary. The exhibition consists of exhibits. For the prototype and experiment, 26 of the 42 exhibits were available. For each exhibit at least four questions for the middle part are available. When users visit the same exhibit, they always get different questions about the exhibit. Because there are a lot of exhibits and each item on the table is duplicated for the amount of users, there is not enough space to map the exhibits one on one to the items. To solve this, the exhibits are divided in categories. My World, Your World is originally divided in eight zones: Stone, Bones, Sea, Human, Sun, Battle, Religion and Art. For the table a minimum of twelve categories is required. A user has to choose six items (which represents categories) and at least six other items should be available for the inverse condition. This means that the zones cannot be used. With the help of Museon employees a new division is made. The available exhibits were taken as the base. The goal was to assign the exhibits to at least twelve categories. The following criteria were used.

• All parts of the exhibition are represented, except some of the rooms that were not in the experi-ment.

• Every category contains at least two categories. This satisfies that when a user picks six items, twelve related exhibits are available.

• The meaning of a category is clear to the user. Even for children.

(31)

Figure 3.6: Multi-touch table language select screen. when he or she sees an image of the category.

With these criteria the exhibits were divided in categories. When the categories were too specific or too global, the procedure was carried out again, till at least twelve, clear categories remained. The categories used were Stones, In the water, On land, In the air, Sea, Prehistory, History, Energy, Religion, Art, Water and Climate. In Appendix F an overview is given of which exhibits and categories were related. Table 3.1 shows the images that represent the categories. Descriptions are given for the categories, because it is hard to cover the content of a category in just a single title.

So the user selects six of the twelve categories. For each selected category two corresponding exhibits are selected, and from each exhibit a question is selected. This results in twelve questions for an user. The questions are ordered according to the structure of the exhibition. This is done to avoid that the users have to go criss-cross through the museum.

Like the language and group screen, the scene shows something fun at every touch. In this scene colored, moving lines are displayed after touching an empty space.

Introduction scene In Figure 3.11 a screenshot of the explanation for the middle part is shown. The

explanation tells the users that they have selected twelve items that they found interesting and that each get twelve questions about those items. The users are told where to go next and encouraged to get a high score. At last a reminder is given that they can work together.

In this scene another way of presenting text is tested. Instead of one page containing the text, each user gets a page with the explanation. To make it a little bit personal, the text starts with the name of the user.

Select items When all users have finished the middle part, they can scan their markers at the group

scene. When this is done, they are redirected to the select item screen, which is shown in Figure 3.12. During the quest, the users have collected twelve personal items each. For the final game they have to pick twelve items in total. The items are grouped in boxes of three or four items, according to the amount of players. There are twelve boxes in total, and from each box one item must be picked. This can be done by touching an item. The selected item gets a green border and the other items in the box become red.

(32)

Figure 3.7: Multi-touch table group screen.

contribute to the final game to get a good result. For a group of four players, there are three boxes for each person.

After all boxes have a selected item, two sliders appear. This sliders can be used to start the final game.

Final game The final game is used to test the gained knowledge of the users. In the middle of the

screen the twelve selected items from the select item scene are placed. Around these items twelve boxes with keywords can be found. From each item a unique related keyword is placed on one of the boxes. The idea of the final game is that users drag a line from a keyword to a related item. The items and keywords can be found in Appendix E. This way they create a simple mind map of items from the exhibition. For each correct connection the group earns two points, but for each incorrect connection they lose one point. And a connection cannot be removed, so users have to be careful before they make a connection. The game has a time limit of 120 seconds. Not every user has seen all the items in the museum, so cooperation is necessary. A screenshot of the final game can be found in Figure 3.13.

Game feedback After the 120 seconds, the connections are checked. This is done by an animation.

Each item is selected one by one. Next all connections from the item are highlighted. A good connection turns green, an incorrect one becomes red. This is shown in Figure 3.14. The connection check animations are shown to give the users feedback on their work. They can see what they have done correct and what incorrect. This should help in the learning process.

At last the final score of the quest is given. The group is congratulated and the final game score, the amount of correct connections, the amount of incorrect connections and the total quest score are shown.

3.2.3

Limitations

(33)

Figure 3.8: Multi-touch table explain screen.

3.3

Multi-touch guidelines in the quest

An overview of multi-touch guidelines is given in Section 3.1. Section 3.2 describes the prototype. This section concludes the chapter by linking the guidelines to the prototype.

MT1: Give users a private workspace in front of them.

The boxes for the tickets are located in the middle of each of the four edges of the table, as can be seen in Figure 3.7. This way each user is restricted to take place at one of those positions. Private user interface elements, for example the name boxes in the account scene, are placed around the ticket boxes. This and the physical distances between the users create private workspaces.

MT2: Place global items in the center of the table.

In several scenes shared elements are used. In the account scene the letters are shared and in the preference and final game scene the items are shared. All these items are located in the center of the table. The letters in the account scene are placed at fixed locations, so they will stay in the center of the table. The items in the preference scene collide with each other, which causes them to move freely around the table. This should not be a problem because each item has enough instances for all users. In the final game the items are located in the center of the table. Users can drag them, but an invisible border prevents them to drag items to the edges. In the group scene the button to start the game can be found at the middle of the table. This button affects all the users, because it confirms the group decision to start a game. Finally the virtual pages used for explanation are initially in the middle, because the explanation affects all.

MT3: Do not require children to drag items over a long distance, without the option to let them lift their fingers.

(34)

Figure 3.9: Multi-touch table account screen.

MT4: Make clear that the table can be used simultaneously.

This point is important at the two start screens, the language select scene and the group scene. When users arrive at the table, only these two screens are available. At these screens the table gives visible feedback at every touch. Users can see that it is possible to interact simultaneously. At the group scene the user is instructed to place the ticket in the ticket box. When this is done, a slider appears that has to be dragged towards the center to join the game. This is done for every single user, so if they follow their own instructions, they find out that they can drag the slider at the same time.

MT5: Use a physical border around the table, so that users can lean without touching. For tables without physical border, be careful with using the edges of the table.

The table did not have a physical border. To cover for this,